Toward high-performance computational chemistry: II. A scalable self-consistent field program

Abstract

We discuss issues in developing scalable parallel algorithms and focus on the distribution, as opposed to the replication, of key data structures. Replication of large data structures limits the maximum calculation size by imposing a low ratio of processors to memory. Only applications which distribute both data and computation across processors are truly scalable. The use of shared data structures that may be independently accessed by each process even in a distributed memory environment greatly simplifies development and provides a significant performance enhancement. We describe tools we have developed to support this programming paradigm. These tools are used to develop a highly efficient and scalable algorithm to perform self‐consistent field calculations on molecular systems. A simple and classical strip‐mining algorithm suffices to achieve an efficient and scalable Fock matrix construction in which all matrices are fully distributed. By strip mining over atoms, we also exploit all available sparsity and pave the way to adopting more sophisticated methods for summation of the Coulomb and exchange interactions. © 1996 by John Wiley & Sons, Inc.

Thread-level parallelization and optimization of NWChem for the Intel MIC architecture

“…The Fock matrix construction is the core computational operation of the widely used Hartree-Fock (HF) method [11,12,9] which is a fundamental approach for solving the Schrö-dinger equation in quantum computing and is often used as the starting point for the accurate but more time consuming electronic correlation methods such as the coupled cluster approach we described in Section 4. Efficient parallelization of the Fock matrix construction can be much more challenging than optimizing CCSD(T).…”

Programmability of the HPCS Languages: A Case Study with a Quantum Chemistry Kernel (Extended Version)

“…Use of the GA library enabled the first scalable fully distributed Hartree-Fock implementation [16,19,25]. The essence of the algorithm can be summarized as follows:…”

Shared memory mirroring for reducing communication overhead on commodity networks

“…The arrays g_DM and g_K are distributed by atomic blocks (as described in [19]), thereby reducing the amount of communication. However, the resulting algorithm is still sensitive to the communication bandwidth.…”